Combustion Tuning for a Gas Turbine Power Plant Using Data-Driven and Machine Learning Approach

2021 ◽  
Vol 143 (3) ◽  
Author(s):  
Suhui Li ◽  
Huaxin Zhu ◽  
Min Zhu ◽  
Gang Zhao ◽  
Xiaofeng Wei
Keyword(s):  
Machine Learning ◽  
Power Plant ◽  
Gas Turbine ◽  
Operating Conditions ◽  
Data Driven ◽  
Ann Model ◽  
Promising Alternative ◽  
Combustion Performance ◽  
Wide Range ◽  
Gas Turbine Power Plant

Abstract Conventional physics-based or experimental-based approaches for gas turbine combustion tuning are time consuming and cost intensive. Recent advances in data analytics provide an alternative method. In this paper, we present a cross-disciplinary study on the combustion tuning of an F-class gas turbine that combines machine learning with physics understanding. An artificial-neural-network-based (ANN) model is developed to predict the combustion performance (outputs), including NOx emissions, combustion dynamics, combustor vibrational acceleration, and turbine exhaust temperature. The inputs of the ANN model are identified by analyzing the key operating variables that impact the combustion performance, such as the pilot and the premixed fuel flow, and the inlet guide vane angle. The ANN model is trained by field data from an F-class gas turbine power plant. The trained model is able to describe the combustion performance at an acceptable accuracy in a wide range of operating conditions. In combination with the genetic algorithm, the model is applied to optimize the combustion performance of the gas turbine. Results demonstrate that the data-driven method offers a promising alternative for combustion tuning at a low cost and fast turn-around.

Combustion Tuning for a Gas Turbine Power Plant Using Data-Driven and Machine Learning Approach

2020 ◽  
Author(s):  
Suhui Li ◽  
Huaxin Zhu ◽  
Min Zhu ◽  
Gang Zhao ◽  
Xiaofeng Wei
Keyword(s):  
Machine Learning ◽  
Power Plant ◽  
Gas Turbine ◽  
Operating Conditions ◽  
Data Driven ◽  
Ann Model ◽  
Promising Alternative ◽  
Combustion Performance ◽  
Wide Range ◽  
Gas Turbine Power Plant

Abstract In this paper, we present a cross-disciplinary study on the combustion tuning of an F-class gas turbine that combines machine learning with physics understanding. An artificial-neural-network-based (ANN) model is developed to predict the combustion performance (outputs), including NOx emissions, combustion dynamics, combustor acceleration, and turbine exhaust temperature. The inputs of the ANN model are identified by analyzing the key operating variables that impact the combustion performance, such as the pilot and the premixed fuel flow, and the inlet guide vane angle. The ANN model is trained by field data from an F-class gas turbine power plant. The trained model is able to describe the combustion performance at an acceptable accuracy in a wide range of operating conditions. In combination with the genetic algorithm, the model is applied to optimize the combustion performance of the gas turbine. Results demonstrate that the data-driven method offers a promising alternative for combustion tuning at a low cost and fast turn-around.

Using Machine Learning to Increase Model Performance for a Gas Turbine System

2020 ◽  
Author(s):  
Samuel M. Hipple ◽  
Zachary T. Reinhart ◽  
Harry Bonilla-Alvarado ◽  
Paolo Pezzini ◽  
Kenneth Mark Bryden
Keyword(s):  
Machine Learning ◽  
Power Plant ◽  
Gas Turbine ◽  
Clean Energy ◽  
Learning Model ◽  
Operating Conditions ◽  
Plant Performance ◽  
Outlet Temperature ◽  
Machine Learning Model ◽  
Critical Components

Abstract With increasing regulation and the push for clean energy, the operation of power plants is becoming increasingly complex. This complexity combined with the need to optimize performance at base load and off-design condition means that predicting power plant performance with computational modeling is more important than ever. However, traditional modeling approaches such as physics-based models do not capture the true performance of power plant critical components. The complexity of factors such as coupling, noise, and off-design operating conditions makes the performance prediction of critical components such as turbomachinery difficult to model. In a complex system, such as a gas turbine power plant, this creates significant disparities between models and actual system performance that limits the detection of abnormal operations. This study compares machine learning tools to predict gas turbine performance over traditional physics-based models. A long short-term memory (LSTM) model, a form of a recurrent neural network, was trained using operational datasets from a 100 kW recuperated gas turbine power system designed for hybrid configuration. The LSTM turbine model was trained to predict shaft speed, outlet pressure, and outlet temperature. The performance of both the machine learning model and a physics-based model were compared against experimental data of the gas turbine system. Results show that the machine learning model has significant advantages in prediction accuracy and precision compared to a traditional physics-based model when fed facility data as an input. This advantage of predicting performance by machine learning models can be used to detect abnormal operations.

A Sequential Approach for Gas Turbine Power Plant Preventative Maintenance Scheduling

2005 ◽  
Vol 128 (4) ◽  
pp. 796-805 ◽  
Author(s):  
Yongjun Zhao ◽  
Vitali Volovoi ◽  
Mark Waters ◽  
Dimitri Mavris
Keyword(s):  
Power Plant ◽  
Gas Turbine ◽  
Preventive Maintenance ◽  
Combined Cycle ◽  
Operating Conditions ◽  
Maintenance Scheduling ◽  
Plant Performance ◽  
Optimization Approach ◽  
Sequential Approach ◽  
Gas Turbine Power Plant

Traditionally, gas turbine power plant preventive maintenance schedules are set with constant intervals based on recommendations from the equipment suppliers. Preventive maintenance is based on fleet-wide experience as a guideline as long as individual unit experience is not available. In reality, the operating conditions for each gas turbine may vary from site to site and from unit to unit. Furthermore, the gas turbine is a repairable deteriorating system, and preventive maintenance usually restores only part of its performance. This suggests a gas turbine needs more frequent inspection and maintenance as it ages. A unit-specific sequential preventive maintenance approach is therefore needed for gas turbine power plant preventive maintenance scheduling. Traditionally, the optimization criteria for preventive maintenance scheduling is usually cost based. However, in the deregulated electric power market, a profit-based optimization approach is expected to be more effective than the cost-based approach. In such an approach, power plant performance, reliability, and the market dynamics are considered in a joint fashion. In this paper, a novel idea that economic factors drive maintenance frequency and expense to more frequent repairs and greater expense as equipment ages is introduced, and a profit-based unit-specific sequential preventive maintenance scheduling methodology is developed. To demonstrate the feasibility of the proposed approach, a conceptual level study is performed using a base load combined cycle power plant with a single gas turbine unit.

scholarly journals One Step Ahead Adaptive Control for Gas Turbine Power Plants

1999 ◽  
Author(s):  
S. M. Camporeale ◽  
L. Dambrosio ◽  
B. Fortunato
Keyword(s):  
Control System ◽  
Power Plant ◽  
Gas Turbine ◽  
Linear Model ◽  
Transfer Functions ◽  
Control Technique ◽  
Steady State Condition ◽  
Wide Range ◽  
One Step ◽  
Gas Turbine Power Plant

The feasibility of the application of One Step Ahead Adaptive (OSAA) Control technique to a gas turbine power plant is investigated. The OSAA technique is a control algorithm especially suitable for non-linear and time-varying systems. This technique uses the Least Square algorithm to estimate in real-time a linear model of the controlled system, and, uses the estimated linear model to evaluate the feedback control variables. The proposed technique allows to control the Gas Turbine power plant in a wide range of electric loads due to its intrinsic adaptive capabilities. Moreover, the OSAA control does not require the knowledge of the dynamic characteristics (e.g. state space systems or transfer functions) in order to design the control system. The OSAA control system has been applied to a single shaft Gas Turbine power plant, which is numerically simulated. The proposed control technique has been tested both in Single-Input Single Output (SISO) mode and in Multi-Input Multi-Output (MIMO) mode. Starting from a steady-state condition, the power plant has been supposed to undergo a step reduction of the electric load. The results show that the OSAA control technique effectively counteracts the load reduction with limited overshoots in the controlled variables and, introducing a integral correction, a negligible static error.

scholarly journals Using Machine Learning Tools to Predict Compressor Stall

2020 ◽  
Vol 142 (7) ◽  
Author(s):  
Samuel M. Hipple ◽  
Harry Bonilla-Alvarado ◽  
Paolo Pezzini ◽  
Lawrence Shadle ◽  
Kenneth M. Bryden
Keyword(s):  
Machine Learning ◽  
Power Systems ◽  
Gas Turbine ◽  
Clean Energy ◽  
Operating Conditions ◽  
Learning Tools ◽  
Power Sources ◽  
Load Following ◽  
Wide Range ◽  
Compressor Surge

Abstract Clean energy has become an increasingly important consideration in today’s power systems. As the push for clean energy continues, many coal-fired power plants are being decommissioned in favor of renewable power sources such as wind and solar. However, the intermittent nature of renewables means that dynamic load following traditional power systems is crucial to grid stability. With high flexibility and fast response at a wide range of operating conditions, gas turbine systems are poised to become the main load following component in the power grid. Yet, rapid changes in load can lead to fluid flow instabilities in gas turbine power systems. These instabilities often lead to compressor surge and stall, which are some of the most critical problems facing the safe and efficient operation of compressors in turbomachinery today. Although the topic of compressor surge and stall has been extensively researched, no methods for early prediction have been proven effective. This study explores the utilization of machine learning tools to predict compressor stall. The long short-term memory (LSTM) model, a form of recurrent neural network (RNN), was trained using real compressor stall datasets from a 100 kW recuperated gas turbine power system designed for hybrid configuration. Two variations of the LSTM model, classification and regression, were tested to determine optimal performance. The regression scheme was determined to be the most accurate approach, and a tool for predicting compressor stall was developed using this configuration. Results show that the tool is capable of predicting stalls 5–20 ms before they occur. With a high-speed controller capable of 5 ms time-steps, mitigating action could be taken to prevent compressor stall before it occurs.

A Sequential Approach for Gas Turbine Power Plant Preventive Maintenance Scheduling

2005 ◽  
Author(s):  
Yongjun Zhao ◽  
Vitali Volovoi ◽  
Mark Waters ◽  
Dimitri Mavris
Keyword(s):  
Power Plant ◽  
Gas Turbine ◽  
Preventive Maintenance ◽  
Gas Turbines ◽  
Combined Cycle ◽  
Operating Conditions ◽  
Maintenance Scheduling ◽  
Plant Performance ◽  
Optimization Approach ◽  
Gas Turbine Power Plant

Traditionally the gas turbine power plant preventive maintenances are scheduled with constant maintenance intervals based on recommendations from the equipment suppliers. The preventive maintenances are based on fleet wide experiences, and they are scheduled in a one-size-fit-all fashion. However, in reality, the operating conditions for each gas turbine may vary from site to site, and from unit to unit. Furthermore, the gas turbine is a repairable deteriorating system, and preventive maintenance usually restores only part of its performance. This suggests the gas turbines need more frequent inspection and maintenance as it ages. A unit specific sequential preventive maintenance approach is therefore needed for gas turbine power plants preventive maintenance scheduling. Traditionally the optimization criteria for preventive maintenance scheduling is usually cost based. In the deregulated electric power market, a profit based optimization approach is expected to be more effective than the cost based approach. In such an approach, power plant performance, reliability, and the market dynamics are considered in a joint fashion. In this paper, a novel idea that economics drive maintenance expense and frequency to more frequent repairs and greater expense as the equipment and components age is introduced, and a profit based unit specific sequential preventive maintenance scheduling methodology is developed. To demonstrate the feasibility of the proposed approach, this methodology is implemented using a base load combined cycle power plant with single gas turbine unit.

scholarly journals Thermodynamic and Economic Evaluation of Gas Turbine Power Plants

2020 ◽  
Vol 7 (1) ◽  
pp. G1-G8
Author(s):  
T. Oyegoke ◽  
I. I. Akanji ◽  
O. O. Ajayi ◽  
E. A. Obajulu ◽  
A. O. Abemi
Keyword(s):  
Power Plant ◽  
Combustion Chamber ◽  
Gas Turbine ◽  
Payback Period ◽  
Operating Conditions ◽  
Plant Performance ◽  
Second Law ◽  
Profitability Analysis ◽  
Second Law Analysis ◽  
Gas Turbine Power Plant

Thermodynamic analysis and economic feasibility of a gas turbine power plant using a theoretical approach are studied here. The operating conditions of Afam Gas Power Plant, Nigeria are utilized. A modern gas turbine power plant is composed of three key components which are the compressor, combustion chamber, and turbine. The plants were analyzed in different control volumes, and plant performance was estimated by component-wise modeling. Mass and energy conservation laws were applied to each component, and a complete energy balance conducted for each component. The lost energy was calculated for each control volume, and cumulative performance indices such as thermal efficiency and power output were also calculated. The profitability of the proposed project was analyzed using the Return on Investment (ROI), Net Present Worth (NPW), Payback Period (PBP), and Internal Rate of Return (IRR). First law analysis reveals that 0.9 % of the energy supplied to the compressor was lost while 99.1 % was adequately utilized. 7.0 % energy was generated within the Combustion Chamber as a result of the combustion reaction, while 33.2 % of the energy input to the Gas Turbine was lost, and 66.8 % was adequately converted to shaft work which drives both compressor and electric generator. Second law analysis shows that the combustion chamber unit recorded lost work of 248.27 MW (56.1 % of the summation), and 77.33 MW (17.5 % of the summation) for Gas Turbine, while air compressor recorded 11.8 MW (2.7 %). Profitability analysis shows that the investment criteria are sensitive to change in the price of natural gas. Selling electricity at the current price set by the Nigerian Electricity Regulation Commission (NERC) at zero subsidies and an exchange rate of 365 NGN/kWh is not profitable, as the analysis of the investment gave an infinite payback period. The investment becomes profitable only at a 45 % subsidy regime. Keywords: energy conversion system, gas turbine, economic analysis, second law analysis, power plant.

Sign in / Sign up

Export Citation Format

Share Document